The Axians Network – Tim Edwards – Director of Technical sales and systems consulting UKI – ADVA – Cloud Expo Europe
The Axians Network – Tim Edwards – Director of Technical sales and systems consulting UKI – ADVA – Cloud Expo Europe
[00:00:18] And welcome back to the Axians network right here at Cloud Expo here at London’s Xcel Centre with water, with water. It has been very dry, very dry Cloud not a very wet time. So that guesstimate, I have to read that you are director of Technical Sales and Systems Consulting UK and Ireland at the time and will continue to adver well known as an optical transmission vendor. But you recently acquired Overset and Ensemble. So is this what I would actually want to share with the thinking behind it? I said I think a change in direction is evolution, evolutionary evolution, I guess.
[00:01:01] So for adver, it’s very much an evolution. So obviously across the industry, we’ve seen a lot of movement towards virtualisation. We’ve seen a lot of movement towards things like software defined networking. Over the past few years, we’ve seen an opening up of interfaces. We’ve seen a lot of disaggregation. We know as a small vendor or a medium sized vendor that what we need to do is we need to be able to interact with a lot of people. We need to work with a lot of other people in a particular way. So when you look at our acquisition of Overture, that was really to add in some functionality that we already started working on, looking at the the open network Egeria. So we are the largest need vendor in the world. We provide an awful lot of hundreds of thousands of Econet needs every year to to to the industry. And what we can do there is with the Internet services. So Maths 2.0 3.0 services. And then what our customers were asking us for was more IP based services on top of that. So you’ve seen an evolution of our needs to provide more capability in that respect. But then there was also this new technology that was coming out called Network Functions virtualisation. And what network functions virtualisation was able to do was to provide another layer on top of that porting in the server blade onto the need to allow us to provide additional services on top. So typically this would be for a UCP environment where you’d have things like firewall, you’d have a router, you’d have a one optimiser. These are the sorts of things that were being asked to put on. And really the idea is there to to collapse down all the all the different boxes that you get at the edge of the network down onto a single endpoint.
[00:02:45] We provide all these different functions and it’s true. Multi vendor isn’t true, is neutral, completely neutral in terms of its supported vendors.
[00:02:52] Absolutely. That’s that’s one that’s one of our real strengths. So if I just come back to the Overture piece, so as a need vendor, we were adding in this this this functionality will coming very much from the hardware. Now, what we saw in Overture was that they were coming at it from a very different direction. So they were coming at it from a very software centric view. So what they were doing was developing a an operating system that could provide MW 2.0 services on servers and also provide a hosting platform for all these virtual functions, which is sitting on top of that on top of a box. So for them, it was very much less about the hardware. It was much more about the software. And then they got the memo around that as well. So they could do the orchestration and they could do. And they got a lot of experience in terms of the different applications that you could you could run on top of these boxes. So bringing these two things together really does give us the best of both worlds, because what we’ve got is an open environment where we’ve got a product called Ensemble Connecter, which is our hosting environment and provides our networking capability. On top of that, we’ve got a very wide ecosystem of partners that we work with, probably in excess of 50 partners now. And we have more partners or more people coming to us than we can actually deal with what we do with those as we on board all of the different virtual network functions on to our vendors. Just to clarify, as the vendors, yes, they come to us. And so we’ve got a service provider, customers who are coming to us asking for product. And that product will be a combination of our UCP software, plus all of the different being offended. So we don’t actually sell the business, but we do the onboarding and we do some level of testing to get them, to get them, to get them working. But equally so. So so what that really does is it allows all the different service providers to have massive amounts of choice. So early on in the industry, what we saw was a lot of vertically integrated stacks. So you would have a network equipment vendor would come along, they would have a piece of hardware on top of that. They would run their own virtual firewall. They would run their own virtual router functionality, and then they would go out and try and sell that to, you know, a service provider who can then offer it as a managed service to the customer. And actually the end customers came back. But what they really said was actually we don’t want to have just a limited set of benefits that are available from one particular manufacturer. What we’d like to do is use our own VNS and this needs to become very much more of an open ecosystem. So we’d like to use check point, we’d like to use Fortinet, whatever. So maybe we shouldn’t drop drop names, but we’ll charge them later. And of course, the other thing that’s come along is, is, is SD-Wan as well. So SD-Wan are traditionally offered as a separate box again at the edge of the network. What we’re doing now is we’re onboarding SD-Wan vendors so you can have a service chain based off a firewall, a router, one optimisation and potentially an SD-Wan vendor as well. So you can get that all together to create this end to end service and, you know, just locating a alone. But now the choice of hardware at the age is very interesting. So Cox is actually probably where you get the best price performance, but there are a number of different areas where sometimes Cox doesn’t really play in. So some of those things are where you need synchronisation. And so that’s that’s one area where servers have the capability. Just yet another area is where it needs to be industrial temperature range. So, you know, where we’ve got you know, we can actually provide hardware that can do both the synchronisation and the timing, industrial temperature range and some 10 gig. And so so some service providers like to have, you know, 10 gig based on vendor grey box, because that’s that’s coming off the shelf is by definition sort of a middle ground that hits the masses and is not a use case for everything.
[00:07:09] So where there are niche points around the kind of main 80, 20 percent rule is there where there are use cases where things are hard and platforms and additional things. So, yeah. So I mean, we know adver very well for for their involvement in core networks. We unified CPS. It’s moving to the edge. It’s moved the access point in the network. I mean, is this this unified CP model move in that way is a deliberate act for yourselves, or is that just the direction the market’s following or so?
[00:07:37] So we’ve always had a play at the edge of the network because of our different needs. But this gives us very much more capability that we’re able to bring to bear at the edge. So, you know, there are lots of other technologies that play in At the Edge. And what UCP is able to give us is it gives us an ability to then connect over multiple different access media when it comes back to the edge. So so we can be DSL, we can be LTE, we can have be less circuits. I’m bringing it all together onto a single UCP box. Then we’re able to provide a lot more a lot more functionality changes the cost of the Kossman as well.
[00:08:15] I mean, I was involved in and some of them NFC and. Designs and solutions three, four or five years ago, and at that point, the kind of model was very much a centralised and virtually used platforms in a data centre and then bring the customers in. And in effect, all the service change was done in that point. But but you have to, in effect, build and they will come. You have to invest heavily in data centre space and that sort of scaled out kind of computing storage to do that. Whereas most of us just don’t want to have that kind of cost. They want to they want to move everything they can into cash flow. So I think moving to the edge, as well as it being a kind of a market direction in terms of customer demand, national also provides sort of trying to push the services that way as well, because then they can infect the customers, sign up. They can actually build that kind of cost out to them rather than doing it themselves and to build an aura around it.
[00:09:10] Yes, definitely. We’re certainly seeing our customers pushing us towards that direction. What’s great about this ensemble connective technology is that it can be both in the data centre and it can also be out at the edges.
[00:09:23] And indeed, what you can do is, is move your different VNS from the data centre to the edge in failure scenarios or where you’ve got an analysis you to try to do some maintenance or something location where they’re best used and most efficient place for you. Absolutely. So there’s no point in having a one optimisation, which is which is two thirds.
[00:09:43] But yeah. So Axians. And that actually has been a problem for many years. We have recently I see a lot more activity. So what’s changed?
[00:09:58] Well, it’s a lot of the kind of points we’ve covered today, really, and we’ve been involved in SD-Wan for three, four years plus we’ve done the one 1.0 SD-Wan 2.0 is coming through the kind of evolution past that is the unified KPA and that model of being able to collapse a whole load of functions into a virtual IoT environment, but drive them from a central point with consistency and have the ability to pop and push. I mean, you have I have models where I have customers going. We like vendor AIX for firewall and they deploy those. They get a new CEO come in. He goes, I hate vendor AIX. I want vendor WI with this model, you could actually do the removal of vendor AIX and the deployment of vendor Y with no truck rolls. And that truck roll is absolutely kills a managed service environment for any kind of margin. They’re going to make our service. So it’s the flexibility that the platforms we’re building into the universal platform we’re building ourselves of Axians we’re going to talk to our customers and then it’s an integral piece of it, just kind of the flexibility in the fact that we have. It is entire vendor agnostic, you know, the one, even if there’s no current support for integration into ensemble. We’ve been talking to the guys, and it’s a relatively short, painless process to get people brought in, so the flexibility there for us and for our customer base is almost infinite. And we’re now starting to look at some of the additional steps up to the platform as a service model that you can build the app. The guys have got into there. They unified CPA, where we can look at bringing third party solutions and we’re looking at bringing models from within the Axians group where we have things Iot sensor controllers virtualise that and put that into the SD-Wan platform with the SD-Wan and a firewall. And then suddenly you’ve got a box that can run an entire Iot yesterday or, you know. The chances and the options are really, really expensive, so it’s quite exciting thing for us and we’re very pleased to be working with the advocates and development indeed.
[00:11:58] Yeah. So flexibility, choice and service philosophy is the other one. So being able to get service out there, change service a DR instead of having to do the truck roll, as you say, that you’re able to just drop that software down there, put it into the service chain, turn up that service chain and the customers back in business drive in centralised IoT functions rather than distribute it.
[00:12:20] So it’s all the cost model and most of the work. Huge volume, huge value. We can talk about this for weeks and weeks. We have one at a time. Have we really?
[00:12:31] Yes, it’s been very good, hasn’t it?
[00:12:35] Thank you very much for your time. Thank you very much. It’s been very interesting. See, the number of people who have seen we’ve seen I mean, we’ve seen, we’ve seen Vinda, we’ve seen yuzu, we’ve seen enabler. And then we have Axians disintegrator. So we have we have the kind of holy Forssmann. I started wrong. I come back so disappointed about how we feel.
[00:12:56] Thank you, Chris, and thank you. This is this has been an Axians production on Disruptive Live here at Cloud Expo in London. So thank you for watching. Thank you.